/

Popular searches

Speech-to-Text

How do I transcribe an audio file?Speech-to-Text overview and models Which ASR model should I use?Streaming ASR over WebSocket ASR request parameters and configuration POST /v1/audio/transcriptions Speech-to-Text Full reference

Text-to-Speech

How do I synthesize text to speech?Text-to-Speech overview Voices, languages, and expression styles TTS streaming WebSocket Audio formats and telephony codecs LLM to TTS pipeline for voice agents Text-to-Speech

API Reference

API Reference hub How do I authenticate API requests?API errors and status codes Rate limits and quotas Request IDs and tracing

Models

Compare all models (capability matrix)ASR model picker, zero-indic, universal, med Open weights on Hugging Face Capability matrix

Features

ASR features and intelligence flags Batch vs streaming, when to use which?Speaker registration and identification APIs Overview What is Shunya?Quickstart Glossary Vāķ Translate BFSI

Intelligence

Intelligence overview with examples Diarization, who spoke when?Intent detection on transcripts Sentiment analysis and summarization Redaction, profanity hashing, PII masking Overview & examples Configuration

Playground Models Log in Free API key

Home Quickstart Speech-to-Text Text-to-Speech Translation Intelligence

ASR API referenceTranscriptions, streaming, speakers TTS API referenceSpeech synthesis, voices, formats Capability matrixModels, languages, pricing

Integrations Deployment

General

API Overview Authentication HTTP error codes API Rate Limits Request IDs

API Reference

POSTPre-Recorded Audio WSSLive Audio GET/health GET/languages

POST/v1/speakers/register GET/v1/speakers/list POST/v1/speakers/identify DELETE/v1/speakers/delete

POSTSingle Text Request WSSContinuous Text Stream GET/health

Docs/API Reference/Request IDs

Request IDs

Every response includes request_id (and an X-Request-Id header). Log it, Shunya support uses it to trace issues. On WebSocket, the session_id in the ready event is the equivalent.

API Rate Limits